Kinetics of Emulsion Polymerization - ACS Publications

the case, diffusion is rapid enough to replace monomer consumed in the ..... small a number of particles because it can be derived from Fick's law tha...
0 downloads 0 Views 4MB Size
1 Kinetics of Emulsion Polymerization Β. M. E. VAN DER HOFF Research and Development Division, Polymer Corp., Ltd., Sarnia, Ontario, Canada

Equations, which are also applicable to suspension,

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

solution, and bulk polymerization, form an exten­ sion of the Smith-Εwart rate theory.

They con­

tain an auxiliary parameter which is determined by the rate of initiation, rate constant of termina­ tion, and volume of the particles. The influence of each variable on the kinetics of emulsion poly­ merization is illustrated.

Two other variables are

the number of particles formed and monomer con­ centration in the particles.

Modifications of the

treatment of emulsion polymerization are required by oil solubility of the initiator, water solubility of the monomer, and insolubility of the polymer in the monomer.

i n few, if any, industrial chemical processes are more "ingredients" used to produce a single final product than i n emulsion polymerization. Moreover, colloidal phenomena play a dominant role i n these polymerization reactions. B o t h these features contribute to the complex nature of emulsion polymerization. It is at present clearly impossible to understand a l l the aspects of these systems. Nevertheless the mechanism a n d kinetics of some emulsion systems are reasonably w e l l understood—those in w h i c h the monomer is water-"insoluble" a n d in w h i c h the polymer is soluble i n the monomer. A n outline is given of this mechanism a n d the kinetics of polymerization are developed on the basis of this mechanism. This theoretical kinetic behavior is then compared w i t h experimental data, both from the literature a n d from unpublished results. Whenever possible, the influence of monomer water solubility a n d monomer solubility of the polymer is commented on. These comments are mostly of a qualitative nature a n d some­ times even speculative. T h e present state of our knowledge does not permit going beyond such comments, although recently the literature has given a few attempts at quantitative interpretation of emulsion polymerization of water-soluble monomers. Emulsion Polymerization Process A brief description is given below of the emulsion polymerization process of a monomer w h i c h is slightly soluble i n water. This picture, w h i c h is n o w gen-

6 In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

VAN

Dik HOFF

Kinetics of Emulsion Polymerization

7

erally accepted, is based on the work done by Harkins (29) and his collaborators during W o r l d W a r I I , published i n 1947. Independently, Yurzhenko and co­ workers developed similar ideas on the mechanism of emulsion polymerization reactions (67, 70). German workers also had reached similar conclusions, but their results were not published i n the open literature [cf. however (22)]. W h e n an emulsifier or soap is dissolved i n water, the solute molecules associ­ ate to form small clusters called micelles. T h e hydrocarbon parts of the emulsifier molecules constitute the interior of the micelles, the surface of w h i c h is formed by the ionic groups of the emulsifier. A small fraction of the soap is molecularly dissolved i n the water and there is a dynamic equilibrium between the micelles and these single molecules i n the aqueous phase. Micelles are of colloidal size, consisting of a relatively small number of soap molecules: of the order of 100 molecules. This corresponds to a diameter of about 50 Α., if one assumes the cluster to be spherical. A t the concentrations usually employed i n emulsion poly­ merization, there are some 1 0 micelles per milliliter of water. 1 8

O n mixing this emulsifier solution w i t h an only slightly water-soluble mono­ mer, a small fraction of the monomer solubilizes i n the micelles—i.e., some monomer dissolves i n the hydrocarbon interior of the micelles, w h i c h swell to roughly double their original size. T h e remainder of the monomer is dispersed i n small droplets, the size of w h i c h depends on the intensity of agitation. T h e diameter of these droplets is usually not smaller than about 1 micron ( 10,000 A . ) and, hence, there are at most some 1 0 droplets per milliliter of water at the normally employed ratio of monomer to water phase. If now an initiator—potassium persulfate, for example—is added to the e m u l ­ sion, it decomposes at sufficiently high temperatures into two radical ions: S 0 * . A t 50° C . , some 1 0 radicals are formed per milliliter of water per second at the usually applied concentration of initiator. T h e radical ions react w i t h what little monomer is dissolved i n the aqueous phase, forming organic sulfate ions. After reaction w i t h a few monomer molecules, the product is a surface active radical ion w h i c h soon becomes incorporated i n a micelle, because of the dynamic charac­ ter of the equilibrium between micellar and molecularly dissolved emulsifier. In the interior of the micelle, the organic radical end of the newly formed chain encounters a high concentration of solubilized monomer and therefore chain growth proceeds rapidly. In the polymerization of styrene at 50° C . , for example, a micelle being "stung" b y a radical expands i n 1 minute to about 250 times its original volume. A l l the monomer required for this almost explosive growth was not contained i n the original micelle, but diffuses to it from the monomer droplets through the aqueous phase. T h e growth of a "stung" micelle is accompanied by the formation of new polymer-water interface onto w h i c h emulsifier molecules are adsorbed from the aqueous phase. Thus, the equilibrium between molecularly dissolved soap and micellar soap is disturbed and micelles disintegrate to restore the concentration of molecularly dissolved emulsifier to its equilibrium value. Thus, i n the process of formation of one polymer particle many micelles disappear. O n l y a small fraction of the original micelles is transformed into polymer particles and the final latex contains some 1 0 particles per milliliter of water. This process of particle formation results i n adsorption of the emulsifier onto polymer particles early i n the reaction; after about 10 to 2 0 % conversion no micelles exist and hence no particles are formed. Polymerization inside a micelle or polymer particle does not deplete the particle of monomer. Because of osmotic forces, the amount of monomer diffusing from the droplets through the water to the particles is i n excess of the amount consumed b y polymerization. 1 1

4

1 3

1 5

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

8

ADVANCES IN CHEMISTRY SERIES

Provided that the monomer droplets are sufficiently small, w h i c h is usually the case, diffusion is rapid enough to replace monomer consumed i n the reaction and to supply the additional amount required for equilibrium swelling of the polymer particles (24). Because of this swelling, monomer droplets disappear from the reaction mixture before polymerization is completed. F r o m this moment on the reaction slows down. F i n a l conversions of close to 1 0 0 % can usually be obtained. This picture of emulsion polymerization mechanisms, mainly due to Harkins (29, 3 0 ) , is generally accepted as is borne out b y its description i n several recent textbooks on polymerization (4, 10, 13, 24, 68). A n extensive review has been given b y Gerrens ( 2 6 ) .

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

Emulsion Polymerization Kinetics T h e process of polymerization consists i n general of three steps: initiation, propagation, a n d termination. I n radical polymerization, a catalyst is usually employed as a source of free radicals, the primary radicals. A fraction of these initiate a rapid sequence of reactions w i t h monomer molecules, the primary radical thus growing into a polymer radical. R a d i c a l activity is destroyed b y reaction of two radicals to form one or two molecules. This termination reaction is called mutual recombination, if only one molecule is formed. Termination b y disproportionation results i n two molecules. F o r many common monomers, recombination is the normal mode of termination and the kinetic treatment here is based on this termination reaction. O n l y slight modifications are required for polymerizations in w h i c h termination occurs b y disproportionation. If both termination processes occur, another variable must be introduced to describe the kinetics of the system fully. L e t Ri designate the rate of initiation of polymer radicals—i.e., the number of gram radicals initiating chains per liter of reaction volume per second ( R mole l i t e r sec." ) . i ?

-1

1

Ri

R. + M

>

RM

r

R a n d k represent the rate a n d the rate constant of the propagation reaction (R , mole l i t e r s e c . ; k , liter m o l e s e c . ) : p

p

-1

p

-1

-1

p

-1

k

M.

+ Μ

n

^->Μ · η

+1

T h e rate and rate constant of the termination process are indicated b y R a n d k (R , gram radicals l i t e r s e c . ; k , liter m o l e s e c . ) : t

-1

t

-1

-1

t

t

-1

k

t

M' n

+ M

m

>lM

n+ m

U n d e r certain conditions, w h i c h i n emulsion polymerization are almost always fulfilled ( 5 6 ) , the concentration of radicals i n the reaction mixture [ R ] is constant. Therefore, the rate of initiation must equal the rate of termination: R = R . Since e

{

R

t

= -d[R-]/dt

= 2 kAR-Y

t

0)

it follows that [/?·]=

(Ri/2kt)

112

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

(2)

VAN DM HOFF

9

Kinetics of Emulsion Polymerization

T h e factor 2 i n these two equations arises from the fact that two polymer radicals disappear on mutual termination. T h e rate of polymerization can n o w be expressed as: R

P

= k [M\ [/?·] = (Rik//2k yinM]

(3)

t

p

where [ M ] is the monomer concentration (mole per l i t e r ) . T h e degree of polymerization (DP)—i.e., the average number of monomer units per polymer molecule—is given b y the number of chain propagation steps for each reaction between a pair of polymer radicals—i.e., for each formation of a " d e a d " polymer molecule. Therefore, D P is equal to the rate of chain growth divided by half the rate of chain termination: DP = 2 Rp/Rt = (2k yRikt) [M\

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

(4)

m

p

These relations for rate and degree of polymerization, although based on a relatively simple chemical picture, agree w e l l w i t h experimental data obtained i n solution a n d bulk polymerizations. I n emulsion polymerization, however, a com­ plication arises from the fact that the volume of the reaction mixture is subdivided into a very large number of very small volume elements, the particles, w h i c h are suspended i n water. These particles are so small that they can accommodate only a limited number of polymer radicals at any one time. T h e peculiarities of emulsion polymerization stem from this limitation. Polymer radicals, being insol­ uble i n water, are confined to the particle i n w h i c h they are generated. T h u s , a radical i n one particle cannot be terminated b y a radical i n another particle. Therefore, the rate of termination i n emulsion is much lower than that given b y Equation 1. T o derive the kinetic relations w h i c h govern emulsion polymerization we reconsider E q u a t i o n 1, while restricting ourselves for the time being to systems w i t h water-soluble initiators where radicals enter the particle one b y one. W e rewrite this equation: Rt

= k J*± x *M± 2

t

n

υ

ν

in w h i c h η is the number of radicals i n the reaction volume, υ, and N is Avogadro's number. T h i s formulation assumes that each radical c a n react not only w i t h other radicals, but also w i t h itself. A s long as the number of radicals i n volume υ is large, this is a good approximation of the actual case—namely, that each radical can react w i t h each other radical, but not w i t h itself. I n emulsion polymerization, however, the number of radicals i n the volume elements is so small that we must us^ the exact expression: A

V

V

in w h i c h ν is now the volume of a particle. It is evident that the rate of termination a n d therefore the kinetic relations depend on the distribution of the radicals among the particles. T o obtain these relations, one must seek information on this distribution. This can be done b y the following considerations, first put forward by Smith and E w a r t ( 5 8 ) . In emulsion polymerizations, the number of particles containing η radicals, N normally assumes a steady-state value early i n the reaction. Therefore the rate of formation of such particles equals their rate of disappearance. A particle containing η radicals is formed either from a particle w i t h n—1 radicals w h i c h n>

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

10

ADVANCES IN CHEMISTRY SERIES

gains a new radical or from a particle containing η + 2 radicals, losing two radicals by mutual termination. T h e rate of radical gain is given b y : An/At = RiN /N

=

A

RN v {

A

where Ν is the total number of particles per liter of organic phase. loss of pairs of radicals i n a particle containing η radicals is given b y : - / An/At l

= / RiN /N l

2

2

= k n(n -

A

T h e rate of

\)/N v

t

A

Hence the rate of appearance of particles containing η radicals is: ANjAt

=

[R N v)N^ i

A

+

l

\k {n + 2)(n +

i)/N v}N

t

A

n+2

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

Similarly particles containing η radicals are lost by either gaining a radical or losing two radicals. T h e rate of loss is: -AN /At n

=

{RiN v}N A

+

n

{k n(n -

\)/N v}N

t

A

n

T h e steady-state condition requires: {R N v}N _ i

A

n

+

l

{k (n + 2){n + \)/N v)N^ t

A

=

2

\RiN v + k n{n A

t

\)/N v)N A

n

T h e general solution of this recursion relation has recently been given by Stockmayer (60). H i s result can be expressed by the following equations: __ η _ * ~ a/4

h(a) h{a)

a = 4(Ri/2k ) N v t

llt

(5)

A

in w h i c h η is the average number of radicals per particle and J and i i are the Bessel functions of the first k i n d of zero and first order, respectively. It follows from Equation 5 that the radical concentration is given b y : 0

[/?·]

= ή N/N

= z(Ri/2k )

A

t

m

This relation is identical with Equation 2 except for the numerical factor z. sequently Equations 3 and 4 become: R

P

= z(Rik /2k y lM] 2

p

t

i2

(6)

Con­ (7)

and DP = z{2k yRikt)^[M} p

(8)

These equations are perfectly general and therefore are applicable to solu­ tion, bulk, suspension, and emulsion polymerization systems. I n the case of solu­ tion and bulk systems there is, i n effect, only one "particle" w i t h a volume of that of the whole reaction mixture. In emulsion and suspension polymerizations, the reaction mixture is subdivided into a large number of small particles and the influence of the state of subdivision is expressed by the factor z. Figure 1 shows the value of this subdivision factor as a function of parameter a (upper c u r v e ) , from w h i c h it can be seen that a polymerizing system follows solution polymerization kinetics (ζ ^ 1) as long as a is larger than about 10. In emulsion systems, a is usually smaller than 1 and both the rate of reaction a n d the molecular weight of the polymer increase w i t h decreasing a. T h e influence of the three variables R k , and ϋ on parameter a are discussed later. F o r values of a between 1 and 10, the character of the polymerization kinetics is intermediate between that of emulsion and that of solution polymerization. This is the region of suspension or pearl polymerization, where the rate and degree of polymerization are somewhat higher than for reaction i n solution. T h e value iy

t

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

VAN DER HOFF

Kinetics of Emulsion Polymerization

11

α Figure 1.

Subdivision factor ζ as a func­ tion of parameter a

From van der Hoff, Β. M . E . , / . Polymer Set. 33, 487 (1958) COURTESY Journal of Polymer Science of the subdivision factor here is roughly between 3 a n d 1. It can readily be ascertained from series expansion of the Bessel functions i n E q u a t i o n 5 that w h e n a approaches zero, ζ - » 2/a and η - » 1/2. T h e unreal case a = 0, η = 1/2 w i l l be called "ideal emulsion polymerization." If a is 1, 0.4, and 0.1, η is, respectively, 0.558, 0.510, a n d 0.5002, w h i c h shows that for a < 1 , the ideal case is rapidly approached. That the average number of radicals per particle should be about 1/2 c a n also b e seen b y considering that, o n entry of a n e w radical into a very small par­ ticle containing one polymer particle, termination between the t w o radicals occurs after a time lapse very small compared w i t h the time interval between successive entries of n e w radicals under the conditions usually prevalent i n emulsion poly­ merization. Hence, the particle contains either one or no radical and the average number of radicals per particle is 1/2. This simple description was first given b y Smith and Ε wart (58) and w i l l be referred to as the Smith-Ewart rate theory. It follows from η = 1/2 that [Κ·] = N/2N . A

Therefore R

p

= Nk [M)/2N v

a

(9)

and hence, the rate of polymerization for a given number of particles is independ­ ent of initiator concentration. This unusual behavior is depicted i n F i g u r e 2. It has been assumed i n the above derivations that the primary radicals enter the particles one b y one. This assumption is fulfilled for water-soluble initiators a n d probably for oil-soluble catalysts activated b y water-soluble compounds. If, however, an unactivated oil-soluble initiator is employed, w h i c h decomposes into radical pairs w i t h i n the particles, this assumption does not hold. A derivation similar to the one given, but based on the formation of radicals i n pairs, leads to a different recursion relation. W i t h the aid of the mathematical method employed by Stockmayer for the former recursion relation, this latter one can be simply solved to y i e l d the same Equations 6, 7, and 8, where now ζ = tanh (a/A). This In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

12

ADVANCES IN CHEMISTRY SERIES

Figure 2.

Schematic representation of number of radicals per particle at two different rates of entry of single radicals

result was first obtained b y H a w a r d (31) b y a somewhat different mathematical route. Values of tanh ( a / 4 ) are also plotted i n Figure 1 (lower c u r v e ) , w h i c h shows that the subdivision factor and therefore rate a n d degree of polymerization decrease w i t h a. F o r small values of α, ζ approaches a/4. Initiation w i t h o i l soluble catalysts is discussed i n more detail later. T h e monomer concentration i n the particles is nearly constant during part of the polymerization. T h e rate of polymerization i n this region at a particular temperature, therefore, depends only on the number of particles present ( Equation 9 ) . This quantity, so important i n emulsion polymerization, is discussed i n the following section. Number of Particles Because of its overriding importance i n determining the rate of emulsion poly­ merization, the factor discussed first is the number of particles formed. T h e mechanism of particle formation described above has been treated quantitatively by Smith and E w a r t (58) on the basis of some simplifying assumptions. T h e most important of these is that a soap molecule occupies the same interfacial area whether it forms part of a micelle or is adsorbed on the polymer-water interface— i.e., during the period of particle formation, the total surface area of micelles plus polymer particles is constant. R i g i d application of diffusion laws leads unfortunately to as yet unsolved mathematical equations. A n approximate solution is obtained b y making either of two further assumptions a n d the number of particles formed is calculated on the basis of each of these. These computations are not given here; they have been amply described. O n l y the results of the calculations are briefly mentioned. Assumption 1. T h e first additional assumption states that only micelles capture radicals. Since polymer particles undoubtedly can do the same, as is shown b y continuing polymerization after the disappearance of the micelles, the number of particles calculated on this basis is too large. T h i s number, N, is given b y : Ν = 0.53 (ρ/μ) " 2

X (a,S)™

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

(10)

VAN DER HO F F

13

Kinetics of Emulsion Polymerization

in w h i c h ρ = rate of radical production μ = rate of volume increase of a particle a = area occupied by one soap molecule S = amount of soap present 8

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

Assumption 2. T h e second additional assumption states that the radical diffusion flux—i.e., the number of radicals entering a particle per unit surface area per unit time—is independent of the radius of the particle. This w i l l lead to too small a number of particles because i t can be derived from Fick's l a w that, i n symmetrical diffusion into a sphere, the current is proportional to the radius of the particle and therefore the diffusion flux is inversely proportional to that radius. In other words, the smaller the particles the more efficient they are i n capturing radicals. Ν calculated on the basis of this assumption is given b y : Ν = 0.37( /μ)2/δ X (aS.J>i*

(10a)

Ρ

This relation differs from the one above only i n the somewhat smaller numerical factor. W i t h these assumptions two different conversion-time curves are associated. It is easily derived that for the case of assumption 1, the number of particles and therefore the rate increase i n proportion to the square of the reaction time, pro­ vided [ M ] is constant ( 3 ) . A t the conversion. C , where the micelles have just disappeared, a l l particles contain a growing polymer radical a n d the rate is double the equilibrium rate given b y Equation 9. T h e rate after C decays exponentially to the equilibrium value. If assumption 2 holds, the rate increases according to a complex function of reaction time. In F i g u r e 3, examples are given of conversion-time curves w h i c h seem to correspond to the theoretical relations derived on the basis of either assumption. T h e usual course of the reaction is probably intermediate between the two types and may i n addition be modified b y changes i n the monomer concentration i n the particles during this period of transition from micelle to particle. T h e conversion at the end of the period of particle formation has been reported b y several authors w h o determined the disappearance of unabsorbed soap. Table I lists some of these results, w h i c h show that for the usual ratio soap-monomer C falls between 10 and 2 5 % conversion, depending on the rate of initiation. v

N

N

Table I.

Values of Conversion at Which Micelles Disappear, C G. Soap/

Monomer Styrene Styrene Styrene

Soap Laurate Myristate Myristate

Styrene

Myristate

Styrene/isoprene, 1/3 Styrene/isoprene, 1/3 Vinyl chloride

Oleate Myristate Myristyl sulfate

700 G.

Monomer 5 5 8 10 5 4.5



Cm,

%

23 24 20

21 12 14 / ) behavior. There are, however, systems for w h i c h ideal emulsion polymerization practically cannot be achieved. It is nevertheless pos­ sible to describe the kinetics of such systems quantitatively. Recently, Gerrens has obtained values of the propagation and termination rate constants at different temperatures for vinyltoluene and vinylxylene (28). T h e termination rate of polymer radicals of these monomers is so low that even at small rates of initiation in small particles, η is larger than / . F r o m measurements of the reaction rate before and after injection of additional initiator in the polymerizing system it was possible to calculate η both at the original and at the boosted initiation rate w i t h the aid of E q u a t i o n 5. Consistent results were obtained when the additional amount of initiator was varied. F r o m these rate data, the termination rate con­ stant was found to be 1 0 and 17 liters mole- s e e r at 45° C . for vinyltoluene and vinylxylene, respectively. These values are to be compared w i t h ~ 10 for styrene (Table I V ) . 2

1

2

1

3

2

1

1

- 4

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

VAN DER HOFF

Kinetics of Emulsion Polymerization

29

Figure 14. Average number of radicals per particle (n) at low conversions as a function of particle size during polymerization of styrene Ο Polymerization initiated by potassium persulfate • Polymerization initiated by cumene hydroperoxide Δ Catculated from (55) Experimental detaih given in Table VII (62)

Figure 15. Rate of reaction in seeded polymerization of styrene as a function of rate of radical production TEA = triethanolamine Rate of radical production calculated with rate constants from (5) Calculated rate of polymerization corresponding ton — 1/2 Cahuhted with kt = 4.2 X 10* liters mole' sec.' 1

1

Conclusion It seems justified to conclude that the theoretical aspects of emulsion poly­ merization kinetics have to a large extent been developed, at least for water-"inIn POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

30

ADVANCES IN CHEMISTRY SERIES

soluble" monomers. Experimental data satisfactorily support the theories. A re­ cent mathematical advance has enlarged the scope of the original Smith-Ewart rate theory and made it possible not only to define accurately the conditions under w h i c h ideal emulsion polymerization behavior might be expected, but also to treat quantitatively nonideal emulsion reactions. It has now become possible to predict polymerization characteristics whether the system is, or is not, subdivided into particles of colloidal dimensions or of microscopic size. Least understood is the mechanism of particle formation. O n l y i n simple systems w i t h water-"insoluble" monomers is this mechanism known and have its kinetic features been confirmed experimentally. M u c h remains to be done w i t h soluble monomers. Coalescence and coagula­ tion of oligomeric and polymeric molecules and indeed of particles have prevented the quantitative treatment of such systems. It might be useful to use seeded poly­ merizations w h i c h provide a means of separating the processes of particle formation and particle growth. Acknowledgment The author thanks H . Gerrens for permission to use his data. Literature Cited

(1) Allen, P. W., J. ColloidSci.,13, 483 ( 1958 ). (2) Allen, P. W., J. PolymerSci.,31, 207 ( 1958 ). (3) Bakker, J., Ph.D. thesis, Utrecht University, Netherlands, 1952; Philips Research Repts. 7, 344 ( 1952 ). (4) Bamford, C. H . , Barb, W . G., Jenkins, A . D., Onyon, P. F., "Kinetics of Vinyl Polymerization by Radical Mechanisms," Academic Press, New York, 1958. (5) Bartholomé, Ε., Gerrens, Η., Ζ. Elektrochem. 61, 522 ( 1957 ). (6) Bartholomé, Ε., Gerrens, H., Herbeck, R., Weitz, H . M., Ibid., 60, 334 ( 1956 ). (7) Belonovskaia, G . B., Dolgoplosk, Β. Α., Vasiutina, Zh.D., Kulakova, M . N . , Izvest. Akad. Nauk S.S.S.R., Otdel. Khim. 1958, 24. (8) Benough, W . I., Melville, H. L . , Proc. Roy. Soc. ( London ) A249, 455 ( 1959 ). (9) Benson, S. W., North, A . M., J. Am. Chem. Soc. 81, 1339 ( 1959 ). (10) Bovey, F . Α., Kolthoff, I. M., Medalia, A . J . , Meehan, E . J . , "Emulsion Polymeriza­ tion," Interscience, New York, 1955. (11) Breitenbach, J . W., Kolloid-Z. 109, 119 ( 1944 ). (12) Breitenbach, J . W., Edelhauser, H . , Makromol. Chem. 44-46, 196 ( 1961 ). (13) Burnett, G . M., "Mechanism of Polymer Reactions," Interscience, New York, 1954. (14) Burnett, G . M., Trans. Faraday Soc. 46, 772 ( 1950 ). (15) Burnett, G . M., Lehrle, R. S., Proc. Roy. Soc. ( London ) A253, 331 ( 1959 ). (16) Burnett, G . M., Lehrle, R. S., Overnall, D . W., Peaker, F . W., J. Polymer Sci. 29, 417 (1958 ). (17) Cherdron, H., Kunststoffe 50, 568 ( 1960 ). (18) Cherdron, H., Schulz, R. C., Kern, W., Makromol. Chem. 32, 197 ( 1959 ). (19) Dolgoplosk, Β. Α., Tinyakova, Ε. I., Khim. Nauka. i. Prom. 2, 280 ( 1957 ); Gummi u. Asbest. 12, 438, 508, 582, 722 ( 1959 ). (20) Edelhauser, H., Breitenbach, J. W., J. Polymer Sci. 35, 423 ( 1959 ). (21) Evans, C. P., Hay, P. M . , Marker, L., Murray, R. W., Sweeting, Ο. J., J. Appl. Polymer Sci. 5, 39 ( 1961 ). (22) Fikentscher, H., Angew. Chem. 51, 433 ( 1938 ). (23) Fikentscher, H., Herrle, H., Ibid., A59, 179 ( 1947 ). (24) Flory, P. J . , "Principles of Polymer Chemistry ," p. 210, Cornell Univ. Press, Ithaca, Ν. Y. 1953. (25) French, D . M., J. Polymer Sci. 32, 395 ( 1958 ). (26) Gerrens, H., Fortschr.Hochpolymeren-Forsch.1, 234 ( 1959 ). (27) Gerrens, Η., Z. Elektrochem. 60, 400 ( 1956 ). (28) Gerrens, H., Köhnlein, E . , Ibid., 64, 1199 ( 1960 ). (29) Harkins, W . D., J. Am. Chem. Soc. 69, 1428 ( 1947 ). (30) Harkins, W . D., J. Polymer Sci. 5, 217 ( 1950 ). (31) Haward, R. N . , Ibid., 4, 273 ( 1949 ). In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.

Downloaded by MAHIDOL UNIV on February 17, 2015 | http://pubs.acs.org Publication Date: January 1, 1962 | doi: 10.1021/ba-1962-0034.ch001

VAN DER HOFF

Kinetics of Emulsion Polymerization

31

(32) Hay, P. M., Light, J. C., Marker, L., Murray, R. W . , Santonicola, A . T., Sweeting, O. J . , Wepsic, J . G., J. Appl. Polymer Sci. 5, 23 ( 1961 ). (33) Jacobi, B., Angew. Chem. 64, 539 ( 1952 ). (34) Jones, J . H., J. Assoc. Offic. Agr. Chemists 28, 398 ( 1945 ). (35) Klevens, H. B., J. Colloid Sci. 2, 365 ( 1947 ). (36) Kolthoff, I. M., Medalia, A . I., J. Polymer Sci. 5, 391 ( 1950 ). (37) Kolthoff, I. M., Miller, I. K., J. Am. Chem. Soc. 73, 3055 ( 1951 ). (38) Kolthoff, I. M., O'Connor, P. R., Hansen, J. L . , J. Polymer Sci. 15, 459 ( 1955 ). (39) L i g h t , J . C., Santonicola, A . T., Sweeting, O. J . , J. Appl. Polymer Sci. 5, 31 ( 1961 ). (40) Matheson, M . S., Auer, Ε. E . , Bevilacqua, Ε. B., Hart, E. J . , J. Am. Chem. Soc. 73, 1700 (1951). (41) Melville, H. W., Valentine, L., Trans. Faraday Soc. 46, 210 ( 1950 ). (42) Moore, D . E . , Parts, A . G., Makromol. Chem. 37, 108 ( 1960 ). (43) Morton, M., Cala, J . Α., Altier, M. W., J. Polymer Sci. 19, 547 ( 1956 ). (44) Morton, M., Kaizerman, S., Altier, M. W., J. ColloidSci. 9, 300 ( 1959 ). (45) Morton, M., Landfield, H., J. Polymer Sci. 8, 111 ( 1952 ). (46) Morton, M., Salatiello, P. P., Landfield, H., Ibid., 8, 279 ( 1952 ). (47) Nakajima, Α., Yamakawa, H., Sakurada, I., Ibid., 35, 489 (1959 ). (48) O'Donnell, J . T., Mesrobian, R. B., Woodward, A . E . , Ibid., 28, 171 ( 1958 ). (49) Okamura, S., Motoyama, T., Intern. Symposium Macromolecular Chem., p. 82, Montreal, 1961. (50) Okamura, S., Motoyama, T., J. Chem. Soc. ( Japan ), Ind. Chem. Sect. 61, 384

(51) Paluit, S. R., Kunststoffe 50, 513 ( 1960 ). (52) Patsiga, R., Litt, M., Stannett, V . , J. Phys. Chem. 64, 801 ( 1960 ). (53) Robertson, E . R., Trans. Faraday Soc. 52, 426 ( 1956 ). (54) Roe, C. P., Brass, P. D . , J. ColloidSci. 9, 602 ( 1954 ). (55) Roe, C. P., Brass, P. D., J. Polymer Sci. 24, 401 ( 1957 ). (56) Schulz, G . V . , Proc. Intern. Symposium Macromolecular Chem., p. 124, Prague, 1957. (57) Smith, W . V . , J. Am. Chem. Soc. 70, 3695 ( 1948 ). (58) Smith, W . V . , Ε wart, R. H., J. Chem. Phys. 16, 592 ( 1948 ). (59) Staudinger, J. J . P., Chem. Ind. ( London ) 1948, 563. (60) Stockmayer, W . H., J. Polymer Sci. 24, 314 ( 1957 ). (61) Thomas, W . M., Gleason, E . H., Mins, G., Ibid., 24, 43 ( 1957 ). (62) van der Hoff, Β. M. E., Ibid., 33, 487 ( 1958 ). (63) Ibid., 44, 241 ( 1960 ). (64) Ibid., 48, 175 ( 1960 ). (65) van der Hoff, Β. M. E., unpublished results, 1958. (66) Vanderhoff, J. W . , Bradford, Ε. B., Tarkowski, H. L., Wilkinson, B. W . ,J.Polymer Sci. 50, 265 ( 1961 ). (67) Voutzii, S. S.. Zeizema. Μ. Α., Uspekhi Khim. 16, 69 ( 1947 ). (68) Walling, C., ''Free Radicals in Solution" Wiley, New York, 1957. (69) Wiener, H . , J. Polymer Sci. 7, 1 ( 1951 ). (70) Yurzhenko, A . T., Kolechova, M . , Doklady Akad. Nauk S.S.S.R. 47, 354 ( 1945 ). (71) Zimmt, W . S., J. Appl. Polymer Sci. 1, 323 ( 1959 ). R E C E I V E D September 6, 1961.

In POLYMERIZATION AND POLYCONDENSATION PROCESSES; PLATZER, N.; Advances in Chemistry; American Chemical Society: Washington, DC, 1962.